Trace element accumulation and elutriate toxicity in ...

MPB-08295; No of Pages 10 Marine Pollution Bulletin xxx (2016) xxx–xxx

Contents lists available at ScienceDirect

Marine Pollution Bulletin journal homepage: www.elsevier.com/locate/marpolbul

Trace element accumulation and elutriate toxicity in surface sediment in northern Tunisia (Tunis Gulf, southern Mediterranean) Walid Oueslati a, Noureddine Zaaboub b, Mohamed Amine Helali b, Rym Ennouri b, Maria Virgínia Alves Martins c, Amel Dhib d, Francois Galgani e, Monia El Bour b, Ayed Added a, Lotfi Aleya d,⁎ a

Laboratoire des Ressources Minérales et Environnement, Département de Géologie, Faculté des Sciences de Tunis, Université Tunis-El Manar, 2092, Tunisia Laboratoire du Milieu Marin, Institut National des Sciences et Technologies de la Mer, 2025 Salammbô, Tunisia Universidade do Estado do Rio de Janeiro — UERJ, Faculdade de Geologia, Av. São Francisco Xavier, 524, Maracanã, 20550-013 Rio de Janeiro, RJ, Brazil d Université de Bourgogne Franche-Comté, Laboratoire de Chrono-Environnement, UMR CNRS 6249, La Bouloie, F-25030 Besançon Cedex, France e IFREMER, LER/CO, Immeuble Agostini, ZI Furiani, 20600 Bastia, France b c

a r t i c l e

i n f o

Article history: Received 25 February 2016 Received in revised form 30 December 2016 Accepted 31 December 2016 Available online xxxx Keywords: Marine sediment Trace metals Toxicity Bioassays Acid-volatile-sulfide

a b s t r a c t Metal concentrations in sediments were investigated in the Gulf of Tunis, Tunisia, in relation to anthropic activities along the Mejerda River and Ghar El Melh Lagoon, with effluents discharged into the gulf. Distribution of grain size showed that the silty fraction is dominant with 53%, while sand and clay averages are 34 and 12% respectively. Zn concentration increased in the vicinity of the Mejerda River while Pb was at its highest levels at the outlet of Ghar El Mehl Lagoon. Sediment elutriate toxicity, as measured by oyster embryo bioassays, ranged from 10 to 45% abnormalities after 24 h, but no relation was found between metal concentration and sediment toxicity. The AVS fraction that represents monosulfide concentrations in the sediment was higher in the central part of the gulf than in the coastal zone. The results reveal the influence of AVS, TOC and grain size on metal speciation and sediment toxicity. © 2016 Elsevier Ltd. All rights reserved.

1. Introduction The accumulation of trace elements in marine sediments is heavily influenced by continental discharges and effluents largely due to fluctuations in the composition of suspended matter (Helali et al., 2016a, 2016b; Picone et al., 2016; Martins et al., 2015; Zaaboub et al., 2015). Trace metals are not definitely fixed within the sediment and many processes affecting metal concentrations in estuarine and coastal marine sediments have been reported as metal recycling through biological, chemical and physical processes (James, 1978; Carignan and Nriagu, 1985; Helali et al., 2016c). Metal accumulation in sediment may also have toxic effects on marine biotas and flora which an increasing number of studies have attempted to describe via differing approaches (USEPA, 2002; Kennedy et al., 2009; Galgani et al., 2009; Zaaboub et al., 2015; Martins et al., 2015; Helali et al., 2016c, Gimbert et al., 2017). SEM/AVS is an approach usually used to evaluate the toxicity of sediment contaminated with heavy metals. SEM are simultaneously extracted metals liberated from sediment by an HCl attack. AVS are acidvolatile sulfides extracted at the same time. Based on the chemical interactions between SEM and AVS, the ratio of these two parameters is used to assess the potential of metal bioavailability and thus sediment ⁎ Corresponding author. E-mail address: lotfi[email protected] (L. Aleya).

toxicity. When SEM/AVS exceeds the value 1, the sediment is considered toxic because there is not enough AVS to scavenge bioavailable metals (SEM). The coast of northern Tunisia is characterized by important industrial and urban centers with well-developed mining activities producing large quantities of effluent discharges in the region including heavy metals (Jdid et al., 1999; Mlayah et al., 2009; Oueslati et al., 2010a,b; Helali et al., 2013; Zaaboub et al., 2014; Ennouri et al., 2015; Helali et al., 2015) and nutrients (Helali et al., 2016a). Most of this effluent, including both sewage and industrial discharge, flows into the 253-km long Mejerda River which in turn flows into the Gulf of Tunis. In recent decades, extraction of Pb, Zn and Ba from mines located in the Mejerda River catchment area have led to large amounts of discharges (Ayari et al., 2016) posing a serious threat to public health (Abidi et al., 2014) and the marine environment. Organic pollution sources have also been evaluated, with concentrations of polycyclic aromatic hydrocarbons in the gulf sediments found within the 0.01 to 2.6 μg g−1 range (Mzoughi et al., 2010). Geochemical studies have therefore been conducted in the Gulf of Tunis, especially in the prodelta of the Mejerda River (Essonni, 1998; Helali et al., 2013; Zaaboub et al., 2014) to investigate by means of simultaneously extracted metals (SEM)/acid volatile sulfide (AVS) the possible toxicity of metals in the marine waters and sediments. The results suggest that there is enough AVS to scavenge metals, thus avoiding

http://dx.doi.org/10.1016/j.marpolbul.2016.12.076 0025-326X/© 2016 Elsevier Ltd. All rights reserved.

Please cite this article as: Oueslati, W., et al., Trace element accumulation and elutriate toxicity in surface sediment in northern Tunisia (Tunis Gulf, southern Mediterranean), Marine Pollution Bulletin (2016), http://dx.doi.org/10.1016/j.marpolbul.2016.12.076

2

W. Oueslati et al. / Marine Pollution Bulletin xxx (2016) xxx–xxx

their toxicity (Helali et al., 2016c). However, works of Helali et al. (2016b,c), using other geochemical approaches such as chemical fractionation of metals in sediments and suspended matter, showed that some elements and especially Pb and Zn are sufficiently bound to the exchangeable fraction making them relatively bioavailable and therefore potentially toxic. To circumvent this paradox, we (1) used the embryo-toxicity bioassay to evaluate sediment toxicity, and (2) examine possible relationships between the observed toxicity and chemical properties of sediments. 2. Material and methods 2.1. Study site The Gulf of Tunis (Fig. 1) is located on the northeastern coast of Tunisia (36°42′–37°10′ N, 10°15′–11°5′ E) and receives sediments and waters essentially from the Mejerda River, the Khlij Channel, and Ghar El Melh Lagoon (Dhib et al., 2013). The Mejerda catchment contains a large number of lead and zinc mines. Ghar El Melh Lagoon, which is the former mouth of the Mejerda, is surrounded by an industrial zone and is connected to the sea via a channel. The Mejerda River Delta is subject to two directions of prevailing winds that differ according to season: from north to northwest during the winter and east to southeast during the summer (Ben Charrada, 1997). The sea surface currents depend on wind direction and are mostly from north to south. 2.2. Sampling Twenty-one surface sediments were sampled using a Van-Veen grab (0–5 cm) from operations aboard the oceanographic vessel “Hannibal” (Fig. 1). All samples were taken from the northern part of the gulf

between 6 and 75 m depth. Sediments were immediately placed in polyethylene flasks and stored at − 4 °C for immediate or short-term analysis, or conserved at −80 °C for later bioassays.

2.3. Grain size distribution Sediment was sieved to collect the fine fraction (63 μm) using nylon mesh. Grain size (0.01–63 μm) was measured in all samples using a laser granulometer (Mastersizer, 2000). Grain size was determined using Stokes Law to characterize particle size distribution for all surface sediment samples. Sediment was placed on the top of a sieve column for a specific time or until it passed through the sieve at a constant low rate, separating different sediment fractions of clay, silt or sand. Duplicate measurements showed fine fraction percentages were reproducible with an analytical b5%.

2.4. Scanning microscope observations Sediment surface was observed by means of a SEM Type JEOL JSM5400 scanning microscope. Observations were made on total sediment.

2.5. Total organic carbon (TOC) Measurements were carried out on total sediment subsamples (b63 μm) by means of a Perkin Elmer PE 2400 CHN. Samples were decarbonated using 1 M HCl solution and dried at 60 °C. Method analysis was detailed in Zaaboub et al. (2014). Duplicate analyses were generally b8%.

Fig. 1. Sampling sites of surface sediment in the Gulf of Tunis.

Please cite this article as: Oueslati, W., et al., Trace element accumulation and elutriate toxicity in surface sediment in northern Tunisia (Tunis Gulf, southern Mediterranean), Marine Pollution Bulletin (2016), http://dx.doi.org/10.1016/j.marpolbul.2016.12.076

W. Oueslati et al. / Marine Pollution Bulletin xxx (2016) xxx–xxx

2.6. Metal analysis The metals analyzed in this study were selected for the following reasons: (1) studies previously conducted on the Gulf of Tunis and Ghar El Melh Lagoon (also located in the Gulf) (Essonni, 1998, Oueslati et al., 2010a, Helali et al., 2013, Zaaboub et al., 2014) revealed a clear contamination by the mentioned elements and have highlighted the Gulf of Tunis sediment geochemical signature. Indeed, the geochemical background (relative to the crust) is naturally contaminated by these metals; this has been explained by the same authors and by the special geochemical signature of the mining watershed of the Mejerda River which governs the sedimentation in the Gulf subject of this study. (2) The concentration of the majority of these metals in surface sediments cited in this study has increased over the concentration reported by Essonni (1998) two decades ago, as mentioned in the Discussion section in the text. The choice was made with the aim of highlighting the continuous pollution by these metals in the Gulf of Tunis and the effect of this pollution on this marine system. Metals were analyzed on fine fraction (b63 μm) using a commercial wavelength dispersive X-ray fluorescence instrument (BRUKER S4 Pioneer), equipped with an Rh anode X-ray tube (60 kV, 150 mA), three analyzer crystals (OVO-55, LiF 200 and PET), a flow proportional counter for light element detection and a scintillation counter for heavy elements. Quantification was controlled using the spectra plus software connected to the equipment. Five grams of each powdered sample were mixed with 0.5 g of a binder (Hoechst wax C micro-powder) and homogenized in agata mortar. A small, 4 cm diameter, aluminum sample holder was used to obtain an XRF-pellet. Pellets were pressed at 90 bars in a Nannetti hydraulic press for 30 s. The accuracy of the analytical procedures used for heavy metal analysis was checked using the BCR32 certified reference material for (Cd, Ni, Cr, Cu, Zn, Fe and Al) and the BCR40

3

for Pb. Only Zr analysis accuracy was not checked in this study. The average uncertainty of duplicate metal analyses was b 10%. To estimate metal accumulation in surface sediment in the Gulf of Tunis, contamination factors (CFs) were calculated according to Hakanson (1980). This ratio is obtained by dividing the concentration (C) of each metal in the sediment by the baseline or background value (concentration in uncontaminated sediment): CF ¼ C Heavy metals=C Background In this study, we used metals concentrations in the deepest sediment core samples (Essonni, 1998) as background values. 2.7. Acid-volatile sulfide (AVS) The AVS concentration was established using about 10 g of wet sample according to the method described by Vogel, 1989 and Added, 2001. Sulfide was liberated with acid attack using 20 ml 6 N HCl 37% EMSURE® ACS. The sample was digested in a 500 ml gas-tight reaction flask for 30–40 min under continuous bubbled nitrogen at a 20 cm3 min−1 flow rate. Developed H2S was received in a trap flask containing 10 ml of 0.1 N iodine solution. Sulfide reacted with an excess of iodine, and the remaining iodine was then determined by titration with 0.1 N sodium thiosulfate solution. Duplicate measurements showed that concentrations of AVS were reproducible with an analytical precision fluctuating around 10%. 2.8. Elutriate embryo-toxicity tests For oyster embryo-toxicity analysis, the procedure described by His et al. (1999) was used. Tests were performed with elutriates prepared

Fig. 2. Spatial distribution of fine fraction of surface sediment in the northern part of the Gulf of Tunis.

Please cite this article as: Oueslati, W., et al., Trace element accumulation and elutriate toxicity in surface sediment in northern Tunisia (Tunis Gulf, southern Mediterranean), Marine Pollution Bulletin (2016), http://dx.doi.org/10.1016/j.marpolbul.2016.12.076

4

W. Oueslati et al. / Marine Pollution Bulletin xxx (2016) xxx–xxx

Fig. 3. (A) Clay micro aggregation from surface sediment of site 13, (B) carbonate dissolution forms from site 3, (C) framboidal pyrite from surface sediment from site 9.

after 10 h agitation of sediments in filtered reference sea water (4/1 V/ W) sampled offshore near Bastia (Corsica, France) and 5 h decantation to separate elutriates (supernatant) from sediments. The mature genitors (Crassostrea gigas) were carefully cleaned and immerged in filtered reference water at 18 °C for 30 min before a thermal shock (28 °C, 30 min). Specimens emitting gametes were placed in two successive baths of filtered reference water. Fecundation was monitored under a microscope; after dilution, the larvae were then placed in Iwaki microplates (300 larvae/well, final volume of 1 ml elutriate) and cultured at 24 °C ± 1 for 24 h. After incubation, the larvae were fixed in 4% formaldehyde and decanted. Positive control was made of CuSO4 and negative control was performed using reference seawater only. Fewer than 12% abnormalities were retained to validate the reproduction process (His et al., 1999). The abnormality rate was determined on the basis of a count of dead or altered (presence of extravelum, non-perfect D shaped shells) larvae per 100 individuals in each well (two replicates per concentration). Results are given as net percentage of abnormalities (Toxicity-Toxicity of control). It is noteworthy that the elutriate sediment toxicity test was performed after elutriate extraction, using an offshore deep-sea water from Bastia as the reference water. Filtration was made through a 0.22 μm membrane just before analysis through bioassay. Whole sediment samples were mixed with 20 ml of filtered reference water (i.e. 250 g l−1) and first shaken for 8 h, followed by 8 h decantation. The mixture of sediment in other water conditions can extract a part of bioavailable phases of accumulated metals (exchangeable fraction, bound to Fe and Mn oxides and bound to organic matter and AVS).

Furthermore, GAMs can be handled with non-normal distributions of the response variable (Züur et al., 2010). Statistical analysis and graphical display were produced using R.2.12.2 (R Development Core Team, 2011). The R packages “Vegan” (Oksanen et al., 2011) and MASS (Venables and Ripley, 2002) were also used. 3. Results 3.1. Grain size distribution Fig. 2 represents the distribution of the fine fraction (b 63 μm) calculated as a sum of the percentages of silts and clays. Sand contents were low in the deeper zones but higher in the coastal region (b10 m depth)

2.9. Statistical analysis Potential relationships between variables were tested by Spearman's correlation coefficient. Redundancy analysis (RDA) was performed to define the effects of sampling sites (Z1, Z2, Z3) on metals, TOC, AVS and bioassay. Statistical analysis was produced by R 2.15.0 (R Development Core Team, 2011) with “Vegan” (Oksanen et al., 2011). Assessment was undertaken of the due indirect relation between toxicity and trace elements that depends on bioavailability, the relationships between toxicity and TOC, the most bioavailable fraction of AVS associated to accumulated metals in lagoon sediment, using generalized additive models (GAM) (Wood, 2006). GAMs can be considered as a nonparametric generalization of linear regressions and are increasingly used in marine ecosystems to study species relationships with multiple organic and inorganic parameters owing to their flexibility allowing non-linear relationships between the covariates (TOC, AVS and toxicity in this study) and the response variable (i.e. metal concentration).

Fig. 4. Distribution of total organic carbon (TOC) in Mejerda River outlet.

Please cite this article as: Oueslati, W., et al., Trace element accumulation and elutriate toxicity in surface sediment in northern Tunisia (Tunis Gulf, southern Mediterranean), Marine Pollution Bulletin (2016), http://dx.doi.org/10.1016/j.marpolbul.2016.12.076

W. Oueslati et al. / Marine Pollution Bulletin xxx (2016) xxx–xxx

5

Table 1 Sediment characterization and major trace element concentrations in Gulf of Tunis surface sediment.

Samples 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 Average Max Min bcr-32 (certified) BCR-32 (this study) Accuracy (%) a

Sand Depth (%)

Silt (%)

Clay (%)

TOC (%)

AVS (%)

Bio-test (%)

Al (%)

Fe (%)

Pb (μg g−1)

Zn (μg g−1)

Cr (μg g−1)

Cu (μg g−1)

Zr (μg g−1)

Ni (μg g−1)

Cd (μg g−1)

6.6 8 8 6.6 14.3 22.3 24.5 13.4 9.1 41.5 31.7 16.1 36.3 54.5 36.7 35.7 54 21.5 54.5 63.2 60 29.5 60 6.6

15 8 26 7 76 58 76 71 41 74 73 42 71 74 68 22 73 58 60 66 65 53 76 7

7 3 8 3 18 12 18 19 19 18 16 13 17 15 16 3 4 12 9 16 14 12 19 3

0.40 0.72 0.16 0.24 0.46 0.08 0.82 0.80 0.18 1.30 0.61 1.10 0.50 0.50 0.40 1.10 0.90 0.40 0.90 0.56 0.70 0.61 1.30 0.08

0.47 0.59 0.42 0.53 0.66 0.44 0.66 0.31 0.44 0.52 0.44 0.20 0.86 0.8 0.55 0.92 0.55 0.20 0.98 0.88 0.66 0.58 0.98 0.20

18 17 43 45 19 35 38 17 28 10 10 28 27 33 22 14 21 35 13 17 24 24 45 10

5.34 5.84 6.03 5.47 4.98 5.63 5.77 5.80 4.61 5.03 5.15 5.02 4.75 5.32 5.33 6.02 5.69 5.23 5.44 5.54 5.52 5.41 6.03 4.61 0.145

3.20 2.92 3.06 2.94 2.25 2.87 2.59 2.95 2.36 2.55 2.89 2.50 2.37 2.53 2.57 2.59 2.79 2.78 2.93 2.70 2.73 2.72 3.20 2.25 0.080

41 27 31 42 31 48 29 43 107 53 59 42 59 41 50 56 55 45 64 53 58 49 107 27 24.2a

197 267 214 184 145 181 163 212 167 163 200 165 163 161 161 170 190 187 196 179 172 183 267 145 253

191 214 180 178 178 177 241 229 166 158 188 181 160 182 197 200 202 181 213 208 194 191 241 158 257

54 69 58 52 54 55 59 55 53 46 60 58 56 52 42 53 55 60 63 63 55 56 69 42 33.7

486 272 430 279 174 365 202 274 202 194 200 170 200 187 138 348 212 208 341 217 230 254 486 138

69 62 75 66 59 64 65 68 63 59 65 65 60 61 56 60 68 70 68 60 64 64 75 56 34.6

2.04 1.45 1.29 1.43 1.31 1.49 1.64 1.15 1.72 1.30 1.48 0.86 0.77 0.82 0.90 1.13 1.16 1.44 1.26 1.63 0.92 1.30 2.04 0.77 20.8

0.132 0.085 25.5

286

250

32.7

37.1

22.8

8.6

13

2.7

2.9

7.2

9

78 89 66 90 6 31 6 9 40 9 11 45 13 11 16 74 23 30 30 18 21 34 90 6

6.2

5.3

Standard material BCR-40.

with an average of 64% (samples 1, 3, 6, 8, 9 and 12). The highest silt and clay average was revealed in the prodelta central zone with rates of 70 and 15%, respectively (Zaaboub et al., 2014). Scanning Electron Microscopy (SEM) shows a micro aggregation of clays (Fig. 3A) and forms of carbonate dissolution (Fig. 3B). Offshore, facing Ghar El Melh Lagoon, the presence of framboidal pyrite is noted (Fig. 3C).

repartition in the area, though slightly decreased in the central part of the gulf (42 μg g−1). The mean metal concentrations in the Mejerda outlet were ranked as Al N Fe N Zn N Cu N Pb.

3.2. Total organic carbon (TOC) Total organic carbon showed an average of 0.66% in the central zone while the coastal zones were at a lower average concentration of around 0.54%. The highest amount of TOC was observed facing the mouth of the Mejerda River (sample 10) and Ghar El Melh lagoon (sample 12) where the percentages are 1.3 and 1.1%, respectively. The lowest values are in coastal zones (Fig. 4) mainly in the gulf's northern part (Table 1). 3.3. Metals Aluminum is mostly considered to be a natural component of sediment particles (i.e., the alumino-silicate mineral fraction) with an average concentration of 5.41% at the 21 sites located in the gulf's northern sector. For Fe distribution, high percentages were assessed in the littoral zone opposite Ghar El Melh Lagoon and facing the outlet of the Mejerda (samples 1, 2, 3, 4 and 6) with an average of 2.98%. Table 1 shows that the majority of metals exhibit significant lateral variation in the Gulf of Tunis surface sediments. Maximum concentration was at 2.04 μg g−1 for Cd. Nickel and chromium ranged from 56 to 75 μg g−1 and 158 to 241, respectively. There were great variations in Zr concentrations either (138–486 μg g−1). Pb, Zn and Cu were not homogeneously distributed. Pb accumulation was enriched mainly in the zone facing Ghar El Melh Lagoon (107 μg g−1) when Zn exhibited higher amounts (267 μg g−1) opposite the Mejerda River mouth (sample 2, Table 1). Cu concentrations showed a more homogeneous

Fig. 5. Distribution of acid volatile sulfides (AVS) in Mejerda River outlet.

Please cite this article as: Oueslati, W., et al., Trace element accumulation and elutriate toxicity in surface sediment in northern Tunisia (Tunis Gulf, southern Mediterranean), Marine Pollution Bulletin (2016), http://dx.doi.org/10.1016/j.marpolbul.2016.12.076

6

W. Oueslati et al. / Marine Pollution Bulletin xxx (2016) xxx–xxx

3.4. Acid-volatile sulfide (AVS) It appeared that AVS distribution followed the same pattern with an average 22.75 and 111.48 mmol kg−1 in the coastal and central zones, respectively. The AVS high value is present in the central part of the gulf (Fig. 5), characterized by low current and fine sediment fraction accumulation. In the same context, Helali et al. (2016c) showed that AVS concentrations in the core collected at 40 m depth in the Gulf of Tunis (central zone in this study) were higher in comparison with those in

the other two cores collected at 10 and 20 m (Fig. 6). These same authors also concluded that there is sufficient AVS to scavenge metals present in the reactive fraction of the sediment by calculating ΣSEM/ AVS ratios in the three studied cores. 3.5. Elutriate embryo-toxicity tests Results showed that sediment toxicity varied between the coastal and central zones, with percentages of 29 and 22%, respectively. Facing

Fig. 6. Vertical profiles of SEM and SEM/AVS (Helali et al., 2016c).

Please cite this article as: Oueslati, W., et al., Trace element accumulation and elutriate toxicity in surface sediment in northern Tunisia (Tunis Gulf, southern Mediterranean), Marine Pollution Bulletin (2016), http://dx.doi.org/10.1016/j.marpolbul.2016.12.076

W. Oueslati et al. / Marine Pollution Bulletin xxx (2016) xxx–xxx

Fig. 7. Distribution of the toxicity in surface sediments in Mejerda River outlet.

Ghar El Melh Lagoon, there was an increase in toxicity that reaches 45% (Fig. 7). Nevertheless, maximum sediment toxicity was registered offshore from the mouth of the Mejerda (46%) while the lowest toxicity (10% abnormalities) was located in the central zones at stations 25 and 27. 3.6. Statistical analysis Tables 2 and 3 show no correlation of toxicity with metals or granulometry. Considering sampling zones (RDA, F = 1.87, p = 0.2) (Fig. 8), statistical results revealed no correlation between metal concentrations, granulometry, TOC, AVS or bioassay in the metal accumulation zones opposite the Mejerda outlet (Z1), Ghar El Melh Lagoon (Z2) and the gulf's central zone (Z3). AVS, TOC and toxicity, however, led to a significant correlation (Fig. 8). 4. Discussion 4.1. Metal accumulation In the fine fraction (b63 μm), metal accumulation was generally correlated with the silty fraction (Zaaboub et al., 2014). Particle size spectrum was demonstrated to be critical when considering the influence of metal distribution in sediments, mainly in presence of aggregated forms as shown in Fig. 3 (Salomons and Fornster, 1984; Cauwet, 1987; Windom et al., 1989; Fang and Hong, 1999). Finer sediments contain more heavy metals as they have a larger surface-to-volume ratio (Martinčić et al., 1990). Grain sediment distribution in this region has been investigated in previous studies, with numerical simulation revealing that sediment dynamics are heavily dependent on littoral

7

current direction and intensity, generated by winds, waves and tides (Brahim et al., 2015). Aluminum concentrations (Table 1) didn't show significant spatial variation indicating that Al enrichment is not due to anthropogenic activities. For Pb, Zn and Cu, the mean concentrations in sediments are higher than those reported in the Essonni (1998) study (Table 4). Results (Table 1) indicated differences in metal concentration distributions. Indeed, the calculation of mean Pb and Zn concentrations in coastal and offshore sediments showed that there was no significant difference for Pb (50.5 μg g−1 in coastal sediments and 47.8 μg g−1 in the offshore sediments). However, Zn showed a different distribution because this difference is more significant between the two mentioned areas (189 and 175 μg g−1). High levels of metal accumulation (267 for Zn, 69 for Cu and 107 μg g−1 for Pb) in sediment were also found in coastal areas, though restricted to stations near the Mejerda (sample 2) and Ghar El Melh Lagoon outlets which are also influenced by the coastal north-south currents (Brahim et al., 2015). Typically, these metals are related to current or former mining activities located in the Mejerda watershed or near the lagoon (Mlayah et al., 2009). The Cd, Ni, Cr and Zr concentrations were higher than those recorded by Essonni (1998) two decades earlier and which are considered as background values for the present study. Compared to other Mediterranean areas, these values remain higher than those reported by Roussiez et al. (2006) in the Gulf of Lion, and by Buccolieri et al. (2006) in Taranto Gulf. Trace metal concentrations are usually influenced by a wide range of factors including both physical and hydrological characteristics such as the composition of benthos, atmospheric conditions, primary productivity and physico-chemical properties (Helali et al., 2015). Interference from geochemical processes must be taken into account in the remobilization of pollutants and trace metals. Metal accumulation in surface sediment in the Gulf of Tunis assessed through contamination factors (CFs) are presented in Table 4. The contamination factor values were interpreted according to Hakanson (1980) where: CF b 1 indicates low contamination; 1 b CF b 3 is moderate contamination; 3 b CF b 6 is considerable contamination; and CF b 6 is very high contamination. In the Gulf of Tunis, the mean CF values for Zn, Pb and Ni were 1 b CF b 3, indicating “moderate contamination”. The CF values for Cu showed “considerable contamination”, while those for Cd indicated “low contamination”. 4.2. Elutriates and metal associated fraction A relationship between elutriates and metal contamination is difficult to prove as it requires more information and specific metal analyses. However, Zaaboub et al. (2015) and Schintu et al. (2015), in their study of sediment toxicity assessment, explained that sediment toxicity may be due in part to metal bioavailability, the latter defined in this context as the metal reactive phase in which metals can be easily dissolved by redox changes and thus become available for easy absorption by living organisms in the sediment. The metal reactive fraction is usually defined as the concentration of the element in question in the nonresidual fraction. In other words, it is the sum of metal concentrations in the exchangeable fraction, linked to carbonates, Fe/Mn oxides, organic matter and sulfides. We will attempt in this discussion, by identifying all possible variables studied in this investigation and others at the same study site, and analyzing each variable relationship, to demonstrate which parameters or processes are capable of evaluating possible toxicity in the Gulf of Tunis sediments. As explained in the introduction, the accumulated polycyclic aromatic hydrocarbons (PAHs) do not reach

Table 2 Correlations of toxicity with metals.

Toxicity

Al

Fe

Pb

Zn

Cr

Cu

Zr

Ni

Cd

r = 0.13 p = 0.7

r = 0.18 p = 0.573

r = −0.04 p = 0.914

r = 0.11 p = 0.733

r = −0.24 p = 0.444

r = −0.17 p = 0.588

r = 0.27 p = 0.404

r = 0.54 p = 0.075

r = −0.24 p = 0.449

Please cite this article as: Oueslati, W., et al., Trace element accumulation and elutriate toxicity in surface sediment in northern Tunisia (Tunis Gulf, southern Mediterranean), Marine Pollution Bulletin (2016), http://dx.doi.org/10.1016/j.marpolbul.2016.12.076

8

W. Oueslati et al. / Marine Pollution Bulletin xxx (2016) xxx–xxx

4.3. Toxicity, TOC and AVS

Table 3 Correlations of toxicity with granulometry.

Toxicity

Clay (%)

Silt (%)

Sand (%)

r = −0.26 p = 0.417

r = −0.28 p = 0.379

r = 0.36 p = 0.256

Fig. 8. RDA TriPlot depicting association between different parameters and sampling zones (Z1: facing Mejerda outlet; Z2: opposite Ghar El Melh Lagoon; Z3: in the central part of the gulf). Red circles refer to parameters. Eigenvalues of the first two axes are indicated by ʎ1 and ʎ2. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

toxicity levels (Mzoughi et al., 2010). Trace element accumulation, as pertains to toxicity, is explained below. Since measurements are taken from elutriates rather than from whole sediments, the results must be interpreted as the real toxicity rate, based only on bioavailable contaminants (Schintu et al., 2015). Chemical speciation (Helali et al., 2013) of metals in the Gulf of Tunis sediments carried out on the same samples has demonstrated that Zn is bound to three groups of fraction: (1) exchangeable fraction, (2) Fe/Mn Oxides fraction and (3) TOC and AVS fraction. Pb has the same distribution as Zn with the exception that it is absent in the Fe/Mn Oxides fraction (Table 5). According to Helali et al. (2013), Pb is better associated with the exchangeable fraction near the coast (25%) than in offshore sediments (17%). However, in the same mentioned study, Pb associated to organic matter and sulfides varies from 38% in coastal sediments to 47% offshore. The same spatial distribution for Zn was noted but with less significance in comparison with Pb. The three bioavailable fractions in this table are mainly represented by organic matter (TOC) and the AVS fraction. Thus, a relationship should be found between toxicity, TOC and AVS.

Toxicity tests have been used in previous studies of larval growth contamination in coastal areas in the oyster Crassostrea gigas (Geffard et al., 2002; Mamindy-Pajany et al., 2010; Zaaboub et al., 2015). Results have shown the test to be highly effective to this species and that it is thus one of the most suitable for toxicity testing. However, by means of this test, results showed no correlation between sediment toxicity and metal concentrations (Table 2). Moreover, high toxicity, which may not be related to organic chemicals, was found close to the water inputs from the Mejerda and Ghar El Melh Lagoon (Mzoughi et al., 2010). During the study an attempt was made using an RDA TriPlot (Fig. 9) to find a direct relation between toxicity zones and metal accumulation, but no significant differences were found among any of the parameters (metals, granulometry, toxicity), nor among sampling zones (RDA, F = 1.87, p = 0.2). In a previous study by Hansen et al. (1999) it was noted that the heavy metal impact on surface sediments and their potential biological toxicity depend on AVS availability. More recently, Blaise and Férard (2005) and Zaaboub et al. (2015) recommended consideration of AVS concentrations in metal-contaminated sediments and in uptake by biotas and subsequent toxicity responses. Present data indicate an accumulation of AVS corresponding to weak toxicity zones. Trace element presence does not necessarily support sediment toxicity. According to Ogendi et al. (2007), Nasr et al. (2014) and El Zokm et al. (2015) bioavailability is an important parameter influencing metal toxicity based on SEM/AVS ratios. The authors evaluated the toxicity of the studied sediments by means of SEM/AVS suggesting that if this ratio is b1, as in the case of the Gulf of Tunis (Helali et al., 2016c), the sediment is considered non-toxic because there is enough AVS to scavenge bioavailable metals (SEM). However, by adopting the other approach for bioavailability evaluation, which is the metal chemical fractionation discussed above, Helali et al. (2013) in their study on the Gulf of Tunis showed that metals are significantly present (Table 5) in the reactive fraction (exchangeable fraction, organic matter and sulfides, Fe/Mn oxides). Sediment-associated trace elements are generally less toxic than dissolved forms as they are not bioavailable. In addition, the toxicity of sediment-associated divalent metals (Pb, Zn and Cu) is also influenced by the quantity of AVS fixing metals (Di Toro et al., 1990). Ankley et al. (1996), Morse and Luther (1999), Added (2001), Rickard and Morse (2005) and Oueslati (2011) have classified some elements according to their affinity to AVS as follows Cu N Pb N Cd N Zn N Ni. Total organic carbon distribution was associated with hydrodynamic factors, with accumulation in the central part of the Gulf of Tunis (Ben Charrada, 1997). Toxicity is then obviously influenced by the TOC content as proposed by Anderson et al. (1987) as in the northern part of the gulf (Fig. 2) where TOC accumulation corresponds to low toxicity. As for trace element concentrations, TOC distribution appears to reflect local environmental accumulation. According to the generalized additive models (GAM), toxicity is correlated with both TOC (F = 5.49; p = 0.006) and AVS (F = 2.76; p = 0.05) (Fig. 9). GAM shows a strong toxicity correlation with low TOC (maximum toxicity with low TOC). In contrast, GAM shows a significant toxicity correlation with high AVS (maximum toxicity with high AVS).

Table 4 Comparative trace elements in marine sediment and contamination factor calculation.

This investigation Essonni (1998) (Gulf of Tunis, Tunisia)* Roussiez et al. (2006) (Gulf of Lions, France) Buccolieri et al. (2006) (Taranto Gulf, Italy) Contamination factor (CF)

Pb (μg g−1)

Zn (μg g−1)

Cr (μg g−1)

Cu (μg g−1)

Zr (μg g−1)

Ni (μg g−1)

Cd (μg g−1)

49 39 39 57 1.26

183 104 122 102 1.76

191 – 73 85

56 15 22 13 3.72

254 – 77 –

64 57 31 53 1.13

1.3 2.01 0.36 – 0.65

Please cite this article as: Oueslati, W., et al., Trace element accumulation and elutriate toxicity in surface sediment in northern Tunisia (Tunis Gulf, southern Mediterranean), Marine Pollution Bulletin (2016), http://dx.doi.org/10.1016/j.marpolbul.2016.12.076

W. Oueslati et al. / Marine Pollution Bulletin xxx (2016) xxx–xxx Table 5 Spatial variation of metal chemical partitioning from the coast (bold values) to offshore in Gulf of Tunis sediments (Helali et al., 2013). Exchangeable fraction Pb (17–25%) Zn (18–20%) Cu 0%

Organic matter (TOC) and AVS fraction

Fe/Mn oxides fraction

(39–47%) (6–8%) (26–30)

0% (44–47%) (35–37%)

Together, TOC and AVS accounted for 78.1% of variability deviance in toxicity (Fig. 9). Indeed, TOC and AVS allowed fixation of metal contaminants, reducing sediment toxicity. Finally, as stated by Hartl (2010), this confirms the importance of measuring relevant parameters such as TOC and AVS when assessing the fate of contaminants, their bioavailability and bioaccumulation potential. Bioaccumulation refers to the capacity of organisms to absorb and concentrate organic and inorganic contaminants, which may be rare in the environment (useful or indispensable, or undesirable or toxic). Overall, results highlighted the importance of AVS and TOC in evaluating metal bioavailability and toxicity. These are marked in areas affected by the Mejerda River and Ghar El Melh Lagoon. Interactions with organic contaminants must now be evaluated for a better understanding of their effects on marine organisms.

5. Conclusion The main accumulated trace elements were lead, zinc and copper. Subjected to the influences of mining and industrial and urban wastes, metal concentrations in surface sediment increases over time. Surface sediment toxicity investigated by applying bioassay shows a toxicity of 50% near the gulf's principal water inflow points. Measurements were taken from elutriates rather than whole sediments; the hypothesis on toxicity origin was based on bioavailable contaminants only. It appears that exchangeable fraction of the accumulated Zn and Pb in surface sediments induces toxicity in elutriates. In addition, a relationship was observed between toxicity and the greatest quantity of the bioavailable fraction for Pb and Zn, organic (TOC) and the AVS fraction, which, based on correlations using generalized additive models, should be considered as a potential source of toxicity in sediment.

Fig. 9. 3D plot showing the nonlinear relationship between toxicity and both TOC and AVS. Map represents GAM estimations.

9

Acknowledgments This study was made possible by the Tunisian-French Cooperation Project (Laboratory of Mineral Resources and the Environment, Tunis Faculty of Sciences, Institut National des Sciences et Technologies de la Mer, Laboratoire Milieu Marin) (IFREMER, Bastia and Chrono-Environment Laboratory, Besançon, UMR CNRS 6249). We would like to thank all the participants for their active participation and valuable contributions. We would also like to thank the editor Professor Charles Sheppard and the anonymous reviewers for the helpful recommendations on this manuscript. References Abidi, S., Bejaoui, M., Jemli, M., Boumaiza, M., 2014. Qualité des eaux du cours principal de la Medjerda et trois de ses affluents nord. Hydrol. Sci. J. http://dx.doi.org/10.1080/ 02626667.2014.909597. Added, A., 2001. Biogeochemical cycles of Org-C, Tot-N and Tot-S in the sediment of the Ghar El Melh Lagoon north of Tunisia. J. Mar. Syst. 30, 139–154. Anderson, J., Birge, W., Gentile, J., Lake, J., Rodgers, J., Swartz, R., 1987. Biological effects, bioaccumulation, and ecotoxicology of sediment-associated chemicals. In: Dickson, K.L., Maki, A.W., Brungs, W.A. (Eds.), Fate and Effects of Sediment-bound Chemicals in Aquatic SystemsSETAC Special Publication 3. Pergamon Press, NY, pp. 267–296. Ankley, G.T., Ma, D., Pesch, C.E., 1996. Predicting the toxicity of metal contaminated field sediments using interstitial concentration of metals and acid-volatile sulphide normalizations. Environ. Toxicol. Chem. 15, 2080–2094. Ayari, J., Agnan, Y., Charef, A., 2016. Spatial assessment and source identification of trace metal pollution in stream sediments of Oued El Maadene basin, northern Tunisia. Environ. Monit. Assess. http://dx.doi.org/10.1007/s10661-016-5402-4. Ben Charrada, R., 1997. Etude hydrodynamique et écologique du golfe de Tunis. Thèse 3ème Cycle. Univ. Tunis II, Fac. Sci de Tunis (319 p.). Blaise, C., Férard, J.F., 2005. Small-scale Freshwater Toxicity Investigations, Volume 1, Toxicity Test Methods. Springer Publishers, Dordrecht, Netherlands (551 p.). Brahim, M., Atoui, A., Sammari, C., Aleya, L., 2015. Surface sediment dynamics along with hydrodynamics along the shores of Tunis Gulf (north-eastern Mediterranean). J. Afr. Earth Sci. 103, 30–41. Buccolieri, A., Buccolieri, G., Cardellicchio, N., Dell'Atti, A., Di Leo, A., Maci, A., 2006. Heavy metals in marine sediments of Taranto Gulf (Ionian Sea, southern Italy). Mar. Chem. 99, 227–235. Carignan, R., Nriagu, J.O., 1985. Trace metal deposition and mobility in the sediments of two lakes near Sudbury, Ontario. Geochim. Cosmochim. Acta 49, 1753–1764. Cauwet, G., 1987. Influence of sedimentological features on the distribution of trace metals in marine sediments. Mar. Chem. 22, 221–234. Dhib, A., Ben Brahim, M., Ziadi, B., Fourat, A., Turki, S., Aleya, L., 2013. Factors driving the seasonal distribution of planktonic and epiphytic ciliates in a eutrophicated Mediterranean Lagoon. Mar. Pollut. Bull. 74, 383–395. Di Toro, D.M., Mahony, J.D., Hansen, D.J., 1990. Toxicity of 319 cadmium in sediments: role of acid volatile sulfide. Environ. Toxicol. Chem. 9, 1487–1502. Ennouri, R., Zaaboub, N., Fertouna-Bellakhal, M., Chouba, L., Aleya, L., 2015. Assessing trace metal pollution through high spatial resolution of surface sediments along the Tunis Gulf coast (southwestern Mediterranean). Environ. Sci. Pollut. Res. http://dx.doi.org/ 10.1007/s11356-015-5775-x. Essonni, N., 1998. Etude de la dynamique des sels nutritifs et des metaux lourds en relation avec la sedimentologie et hydrodynamique dans le large du golfe de Tunis. PhD thesis. Univ. Tunis II, Fac. Sci de Tunis (229 p.). Fang, T.H., Hong, E., 1999. Mechanisms influencing the spatial distribution of trace 412 metals in surficial sediments off the south-western Taiwan. Mar. Pollut. Bull. 38, 1026–1037. Galgani, F., Senia, J., Guillou, J., Laugier, T., Munaron, D., Andral, B., Guillaume, B., Coulet, E., Boissery, P., Brun, L., Bertrandy, M., 2009. Assessment of the environmental quality of French Continental Mediterranean lagoons with oyster embryobioassay. Arch. Environ. Contam. Toxicol. 57, 540–551. Gimbert, F., Quentin, P.J., Al Ashoor, A., Cretenot, C., Aleya, L., 2017. Encaged Chironomus riparius larvae in assessment of trace metal bioavailability and transfer in a landfill leachate collection pond. Environ. Sci. Pollut. Res. http://dx.doi.org/10.1007/s11356016-8261-1. El Zokm, G.M., Okbah, M.A., Younis, A.M., 2015. Assessment of heavy metals pollution using AVS-SEM and fractionation techniques in Edku Lagoon sediments, Mediterranean Sea, Egypt. J. Environ. Sci. Health A Tox. Hazard. Subst. Environ. Eng. 50-6: 571–584. http://dx.doi.org/10.1080/10934529.2015.994945. Geffard, O., Budzinski, H., His, E., Seaman, M.N.L., Garrigues, P., 2002. Relationships between contaminant levels in marine sediments and their biological effects on embryos of oyster, Crassostrea gigas. Environ. Toxicol. Chem. 21, 2310–2318. Hakanson, L., 1980. An ecological risk index for aquatic pollution control: a sedimentological approach. Water Res. 14, 975–1001. Hansen, D.J., Berry, W.J., Mahony, J.D., Boothman, W.S., Di Toro, D.M., Robson, D.L., Morse, J.W., Luther, G.W., 1999. Chemical influences on trace metal inter-actions in anoxic sediments. Geochim. Cosmochim. Acta 63, 3373–3378. Hartl, M.G.J., 2010. Biomarkers in integrated ecotoxicological sediment assessment. In: Poletto, C., Charlesworth, S. (Eds.), Sedimentology of Aqueous Systems, pp. 147–170. Helali, M.A., Oueslati, W., Zaaboub, N., Added, A., Abdeljaouad, S., 2013. Geochemistry of marine sediments in the Mejerda River delta, Tunisia. Chem. Speciat. Bioavailab. 4, 247–257.

Please cite this article as: Oueslati, W., et al., Trace element accumulation and elutriate toxicity in surface sediment in northern Tunisia (Tunis Gulf, southern Mediterranean), Marine Pollution Bulletin (2016), http://dx.doi.org/10.1016/j.marpolbul.2016.12.076

10

W. Oueslati et al. / Marine Pollution Bulletin xxx (2016) xxx–xxx

Helali, M.A., Zaaboub, N., Oueslati, W., Added, A., Aleya, L., 2015. Diagenetic processes and sediment-water exchanges of heavy metals in the Mejerda River Delta (Gulf of Tunis). Environ. Earth Sci. 74, 6665–6679. Helali, M.A., Zaaboub, N., Oueslati, W., Added, A., Aleya, L., 2016a. Nutrient exchange and oxygen demand at the sediment–water interface during dry and wet seasons off the Mejerda River Delta (Tunis Gulf, Tunisia). Environ. Earth Sci. 75:25. http://dx.doi.org/ 10.1007/s12665-015-4820-x. Helali, M.A., Zaaboub, N., Oueslati, W., Added, A., Aleya, L., 2016b. Chemical speciation of Fe, Mn, Pb, Zn, Cd, Cu, Co, Ni and Cr in the suspended particulate matter off the Mejerda River Delta (Gulf of Tunis, Tunisia). J. Afr. Earth Sci. http://dx.doi.org/10. 1016/j.jafrearsci.2016.02.013. Helali, M.A., Zaaboub, N., Oueslati, W., Added, A., Aleya, L., 2016c. Bioavailability and assessment of heavy metal pollution in sediment cores off the Mejerda River Delta (Gulf of Tunis): how useful is a multiproxy approach? Mar. Pollut. Bull. http://dx. doi.org/10.1016/j.marpolbul.2016.02.027. His, E., Heyvang, I., Geffard, O., De Montaudouin, X., 1999. A comparison between oyster (Crassostrea gigas) and sea urchin (Paracentrotus lividus) larval bioassays for toxicological studies. Water Res. 33, 1706–1718. James, R.O., 1978. Effects of heavy metals on aquatic life. Commonw. Sci. Indus. Res. Org. Canberra. Springer Verlag, Berlin (p. 78 Cited in: Salomons and Förstner 1984, Metals in the Hydrocycle). Jdid, E.A., Blazy, P., Kamoun, S., Guedria, A., Marouf, B., Kitane, S., 1999. Environmental impact of mining activity of the pollution of the Mejerda stream, north-west Tunisia. Bull. Eng. Geol. Environ. 57, 273–280. Kennedy, A.J., Steevens, J.A., Lotufo, G.R., Farrar, J.D., Reiss, M.R., Kropp, R.K., Doi, J., Bridges, T.S., 2009. A comparison of acute and chronic toxicity methods for marine sediments. Mar. Environ. Res. 68, 118–127. Mamindy-Pajany, Y., Libralato, G., Roméo, M., Hurel, C., Losso, C., Ghirardini, A.V., Marmier, N., 2010. Ecotoxicological evaluation of Mediterranean dredged sediment ports based on elutriates with oyster embryotoxicity tests after composting process. Water Res. 44, 1986–1994. Martinčić, D., Kwokal, Ž., Branica, M., 1990. Distribution of zinc, lead, cadmium and copper between different size fractions of sediments II. The Krka River Estuary and the Kornati Islands (Central Adriatic Sea). Sci. Total Environ. 95, 217–225. Martins, M.V.A., Zaaboub, N., Aleya, L., Frontalini, F., Laut, L., Miranda, P., Silva, F., Pereira, E., Mane, M., Rocha, F., El Bour, M., 2015. Environmental quality assessment of Bizerte Lagoon (Tunisia) using living foraminifera assemblages and a multiproxy approach. PLoS One 10 (9), e0137250. Mlayah, A., Ferreira Da Silva, E., Rocha, F., Ben, Hamza., C.H., Charef, A., Noronha, F., 2009. The Oued Mellègue: mining activity, stream sediments and dispersion of base metals in natural environments, North-western Tunisia. J. Geochem. Explor. 102, 27–36. Morse, J.W., Luther, G.W., 1999. Chemical influences on trace metal-sulfide interactions in anoxic sediments. Geochim. Cosmochim. Acta 63, 3373–3379. Mzoughi, N., Chouba, L., Lespes, G., 2010. Assessment of total aromatic hydrocarbons, aliphatic and polycyclic aromatic hydrocarbons in surface sediment and fish from the Gulf of Tunis (Tunisia). Soil Sediment Contam. 19, 467–486. Nasr, S.M., Khairy, M.A., Okbah, M.A., Soliman, N.F., 2014. AVS-SEM relationships and potential bioavailability of trace metals in sediments from the Southeastern Mediterranean Sea, Egypt. Chem. Ecol. 30, 15–28. Ogendi, G.M., Brumbaugh, W.G., Hannigan, R.E., Farris, J.L., 2007. Effects of acid-volatile sulfide on metal bioavailability and toxicity to midge (Chironomus tentans) larvae in black shale sediments. Environ. Toxicol. Chem. 26-2, 325–334.

Oksanen, J., Blanchet, G., Kindt, R., Legendre, P., O'Hara, R., Simpson, G., Solymos, P., Stevens, H., Wagner, H., 2011. Vegan: Community Ecology Package. R Package. Version 1 17-11. Oueslati, W., Added, A., Abdeljaouad, S., 2010a. Evaluation of metal contamination in a changed sedimentary environment: Ghar El Melh Lagoon, Tunisia. Chem. Speciat. Bioavailab. 22-4, 227–240. Oueslati, W., Added, A., Abdeljaoued, S., 2010b. Vertical profiles of simultaneously extracted metals (SEM) and acid-volatile sulfide in a changed sedimentary environment: Ghar El Melh Lagoon, Tunisia. Soil Sed. Cont. 19-6, 696–706. Oueslati, W., 2011. Cycles biogéochimiques des métaux lourds dans les sediments marins de la lagune de Ghar El Melh. Thèse de doctorat, 270 p. Université de Tunis El Manar. R Development Core Team (2011) R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-070, URL http://www.R-project.org/. Picone, M., Bergamin, M., Losso, C., Delaney, E., Arizzi Novelli, A., Ghirardini, A.V., 2016. Assessment of sediment toxicity in the Lagoon of Venice (Italy) using a multi-species set of bioassays. Ecotoxicol. Environ. Saf. 123, 32–44. Rickard, D., Morse, J.W., 2005. Acid volatile sulfide (AVS). Mar. Chem. 97, 141–197. Roussiez, V., Ludwig, W., Monaco, A., Probst, J.L., Bouloubassi, I., Buscaila, R., Saragonia, G., 2006. Sources and sinks of sediment-bound contaminants in the Gulf of Lions (NW Mediterranean Sea): a multi-tracer approach. Cont. Shelf Res. 26, 1843–1857. Salomons, W., Fornster, U., 1984. Metals in the Hydrocycle. Springer Verlag, New York (349 p.). Schintu, M., Buosi, C., Galgani, F., Marrucci, A., Marras, B., Ibba, A., Cherchi, A., 2015. Interpretation of coastal sediment quality based on trace metal and PAH analysis, benthic foraminifera, and toxicity tests (Sardinia, Western Mediterranean). Mar. Pollut. Bull. 94, 72–83. USEPA, 2002. Methods for Measuring the Acute Toxicity of Effluents and Receiving Waters to Freshwater and Marine Organisms. fifth ed. US Environmental Protection Agency, Office of Water, Washington, DC (EPA-821-R-02-012). Venables, W.N., Ripley, B.D., 2002. Modern Applied Statistics with S. fourth ed. Springer, New York (ISBN 0-387-95457-0). Vogel, A.I., 1989. Vogel's Textbook of Quantitative Chemical Analysis. fifth ed. John Wiley & Sons Revised by G. H. Jeffery et al. Windom, H.L., Schropp, S.J., Calder Ryan, J.D., Smith, R.G., Burney, L.C., Lewis, F.G., Rawlinson, C.H., 1989. Natural trace metal concentrations in estuarine and coastal marine sediments of the Southeastern United States. Environ. Sci. Technol. 23, 314–320. Wood, S.N., 2006. Generalized Additive Models: An Introduction with R. Chapman and Hall/CRC. Zaaboub, N., Oueslati, W., Helali, M.A., Abdeljaouad, S., Huertas, J.F., Galindo, A.L., 2014. Trace elements in different marine sediment fractions of the Gulf of Tunis, Central Mediterranean Sea. Chem. Speciat. Bioavailab. 26-1, 1–12. Zaaboub, N., Martins, M.V.A., Dhib, A., Bejaoui, B., Galgani, F., El Bour, M., Aleya, L., 2015. Accumulation of trace metals in sediments in a Mediterranean Lagoon: usefulness of metal sediment fractionation and elutriate toxicity assessment. Environ. Pollut. 207, 226–237. Züur, A., Ieno, E., Walker, N., Saveliev, A., Smith, G., 2010. Mixed effects models and extensions in ecology with R. Statistics for Biology and Health (549 p.).

Please cite this article as: Oueslati, W., et al., Trace element accumulation and elutriate toxicity in surface sediment in northern Tunisia (Tunis Gulf, southern Mediterranean), Marine Pollution Bulletin (2016), http://dx.doi.org/10.1016/j.marpolbul.2016.12.076